15:59:48 <cohosh> #startmeeting anti-censorship meeting
15:59:48 <MeetBot> Meeting started Thu Oct 14 15:59:48 2021 UTC.  The chair is cohosh. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:59:48 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:59:54 <cohosh> welcome :)
16:00:09 <cohosh> here is our meeting pad: https://pad.riseup.net/p/tor-anti-censorship-keep
16:00:38 <meskio> hello
16:00:40 <anadahz> o/
16:01:23 <cohosh> feel free to add to the agenda
16:02:26 <cohosh> first on the discussion is just an update on snowflake usage
16:02:51 <cohosh> it's been a pretty steady ramp up since it was shipped in tor browser stable
16:03:07 <cohosh> and now we're getting closer to meek levels
16:03:09 <cohosh> https://metrics.torproject.org/userstats-bridge-transport.html?start=2021-07-16&end=2021-10-14&transport=meek&transport=snowflake
16:03:48 <dcf1> had anyone heard about a Snowflake-related project by the Calyx Institute?
16:04:37 <cohosh> oh, we're working with calyx on a new grant
16:04:50 <dcf1> I just heard today at anadahz's PT meetup that there will be a lightning talk about it next week
16:05:08 <cohosh> i'm not sure i'm familiar with snowflake work on their side related to that grant though
16:05:28 <dcf1> ok, I was just curious
16:06:22 <meskio> I see the throughput on the snowflake/meek bridges is similar on MB/s but the nubmer of clients is ~1/4 in snowflake, I guess meek has some kind of limit on the traffic it can do because of the price of it or something
16:06:56 <cohosh> yea i was wondering about that as well
16:07:07 <dcf1> yes, afaik the meek bridge is rate limited, or else the bandwidth use would be much higher because of the demand
16:07:07 <cohosh> dcf1: how did the throughput throttling work for meek?
16:07:18 <cohosh> (if you know)
16:07:27 <dcf1> back when I ran a meek bridge I had to keep it rate limited. I don't know what the situation is with the current operators but I imagine it's the same.
16:07:44 <meskio> about calyx, I think they are working with leap on using snowflake in their VPN, not sure how far they are
16:08:09 <dcf1> https://gitlab.torproject.org/tpo/network-health/metrics/timeline has a bunch of related entries, e.g. 2018-05-08 "The meek-amazon bridge imposes a rate limit of 1 MB/s."
16:08:29 <dcf1> cohosh: I think it was just using BandwidthRate in torrc
16:09:00 <cohosh> okay thanks!
16:10:07 <cohosh> anadahz: is the next item about the PT spec yours?
16:12:46 <anadahz> yes
16:13:38 <anadahz> There is a new proposed PT spec (v2.2.1) https://github.com/Pluggable-Transports/Pluggable-Transports-spec/tree/main/releases/PTSpecV2.2.1
16:14:04 <anadahz> Please RFC if you can.
16:14:18 <cohosh> nice!
16:14:23 <anadahz> thx!
16:14:34 <meskio> cool
16:14:54 <cohosh> is there a preferred place to leave comments?
16:15:40 <anadahz> Not really (at least not yet).
16:16:20 <anadahz> I am trying to put the specs in a text file format so that review can be easier and have comments per line.
16:17:00 <dcf1> good idea
16:17:29 <anadahz> Also please let me know if you have any suggestions to help reviewing the specs easier (thx).
16:17:45 <cohosh> yea the txt format will be nice
16:17:59 <meskio> +1
16:18:07 <anadahz> For the time being you can use the issues functionality on GitHub.
16:18:25 <cohosh> great
16:19:15 <anadahz> dcf1: Not sure what is the Calyx lighting talk about, I guess we will find out when the complete program is out.
16:20:07 <cohosh> that's a good segue into the next agenda item about PTIM next week :D
16:20:20 <anadahz> ;)
16:21:19 * cohosh is looking forward to it
16:22:18 * meskio will join it too :)
16:23:05 <cohosh> link: https://www.pluggabletransports.info/ptim2021/
16:23:29 <cohosh> anadahz: anything else you want to add before we move on to the next items?
16:23:40 <anadahz> Registration closes: October 17th 2021 AoE
16:24:23 <meskio> next sunday
16:24:44 <anadahz> (sorry for adding the agenda items in the wrong order, I thought they have been old discussion points)
16:25:05 <anadahz> meskio: this Sunday, in 3 days.
16:25:31 <cohosh> oh no worries! some of them are old, i just wanted to recap to make sure they've been interpreted correctly
16:26:50 <cohosh> okay for the next thing, i opened some MRs for snowflake and the webextension
16:27:08 <cohosh> and i wanted to make sure they match the agreement from last meeting
16:27:17 <dcf1> I will look at those. Thanks for doing it.
16:27:23 <cohosh> thanks!
16:27:40 <cohosh> i let all snowflake.git pieces point to *.torproject.net
16:27:50 <cohosh> and the webextension point to *.freehaven.net
16:28:24 <dcf1> that sounds right
16:28:28 <cohosh> great
16:28:41 <cohosh> okay, the last item for discussion is choosing a reading for reading group
16:29:00 <dcf1> https://github.com/net4people/bbs/wiki/Reading-list
16:31:31 <cohosh> hrm this internet performance in china paper sounds interesting
16:32:03 <meskio> that one? https://censorbib.nymity.ch/#Zhu2020a
16:32:35 <cohosh> yup
16:32:53 <cohosh> and also meteor
16:33:14 <cohosh> well, they all look interesting honestly XD
16:34:13 <cohosh> and the weaponizing middle boxes work
16:34:13 <meskio> both of them sounds interesting
16:35:13 <cohosh> any preferences out of those three?
16:35:54 <meskio> we could start with the cina peformance, is the oldest one, if that is a reason to do it
16:36:01 <meskio> s/cina/china/
16:36:22 <meskio> I don't have a preference
16:37:19 <dcf1> China performance is a good idea
16:37:27 <cohosh> great :D
16:37:35 <cohosh> so, two weeks from now?
16:38:16 <dcf1> All Hallow's Eve Eve Eve Eve
16:38:31 <meskio> nice
16:38:41 <cohosh> :o
16:38:52 <cohosh> lol
16:39:10 <cohosh> okay any more discussion for today or needs help with?
16:39:28 * arma2 checks the recommended costume advice for All Hallow's Eve Eve Eve Eve
16:39:51 <dcf1> business casual, but with black eyeliner
16:39:59 <arma2> scary
16:41:25 <cohosh> oh dcf1 i was interested to see a nonzero number for the azure snowflake accounting
16:41:36 <cohosh> despite us switching the tor browser bridge line to fastly
16:41:42 <dcf1> yes I'm still not sure where that is coming from
16:42:07 <dcf1> we could turn it off and see who complains
16:43:42 <cohosh> fair enough, i know orbot has switched in the code but maybe there are old orbots or old tor browsers kicking around
16:44:09 <dcf1> but it's also nbd either way
16:44:47 <anadahz> dcf1: nbd?
16:44:52 <dcf1> no big deal
16:46:22 * cohosh waits another min
16:47:09 <cohosh> #endmeeting